주택연금의 최적 수요자 계층분석을 통한 장수리스크 완화 효과에 관한 연구

The objective of this study is to examine how much longevity riskcan be alleviated by using the reverse mortgage loan system in Korea(JooTaekYeonKeum, JTYK). In details, we compare expected utility valueof JTYK users during the retirement period with that of non-JTYK users,and try to identify characteristics of the groups which earn most benefitsout of the JTYK in terms of the expected utility value. To achieve this goal,we used data from the 2011 Korean Retirement and Income Study (KReIS)of National Pension Research Institute and extracted relevant demographicdata of the households lead by a male householder, whose age is rangedfrom 60 to 65 years. For each household with a specific age of the headand value of the house its members reside, different monthly payment streamsfrom the JTYK according to four different payment types were obtained,and by using a mathematical model whose objective is to maximize expectedutility function during the retirement period, the maximum expected utilityvalues were calculated by assuming that (1) the household uses the JTYKand (2) the household does not use the JTYK and self-annuitizes the wealth. For analysis, we first define the difference of utility value from the twoassumptions, i.e. the utility value from the assumption (1) -utility valuefrom the assumption (2) as the amount of effect of longevity risk alleviationby using the JTYK. Then, ANOVA (analysis of variance) and multi-variateregression analyses were performed. The results of the ANOVA indicate that the effect of longevity riskalleviation by using the JTYK is significantly different among groups bynet wealth excluding house, income and consumption. The direction of the differences is opposite to the housing cost of non-JTYK users during theretirement period. Many effects in the multi-variate analyses are consistentwith major effects shown in the ANOVA. Specifically, the results of theregression, in which the other factors are controlled, show that the effectof longevity risk alleviation by the JTYK is significantly different as thevalue of house, net wealth except the house value, age of the householder,or public pension changes. Furthermore, the direction of the effect for thecases where the housing cost during the retirement period of non-JTYKusers is no larger than 30% of value of the house is opposite to that forthe cases where the housing cost of non-JTYK users is no less than 50%of value of the house. The results of this study suggest that the housingcost during the retirement be a crucial factor when individuals considerin opting for the JTYK and policy makers try to improve the design of the JTYK.